Neurofibromatosis type 1 (NF1) is a common genetic disorder with a high degree of variability of clinical expression, including skeletal abnormalities in over 1/3 of patients. This disorder is associated with spinal abnormalities, long bone dysplasia, and sphenoid wing dysplasia. These osseous manifestations are unpredictable, and the pathogenesis, natural history, and clinical outcome remain relatively obscure. The spinal abnormalities are varied and include scoliosis (common and dystrophic forms), neurofibromas,dural ectasias, meningoceles, and vertebral defects. The primary objectives of this clinical study are to determine the incidence and clinical history of NF1-related spinal abnormalities in a prepubertal cohort of 120 children with NF1 over 3 years. Secondary objectives are to determine the efficacy of various radiographic screening tools as predictors for dystrophic scoliosis. These goals will be accomplished in 3 specific aims. Aim 1 is to identify associations of spinal cord dural ectasias, spinal neurofibromas,and meningoceles with dysplastic osseous abnormalities and dystrophic scoliosis. Spine radiographs and MRI will be used to test the hypothesis that NF1 patients with certain manifestations are more likely to develop dystrophic scoliosis. Aim 2 is to define the clinical history and short-term outcome of dystrophic scoliosis and spine abnromalities with respect to various radiographic indices. It tests the hypothesis that there are quantitative differences in vertebral scalloping and spinal canal and vertebral body cross-sectional areas when there is an associated additional dystrophic abnormality, and these differences are prognostic indicators for dystrophic scoliosis. Aim 3 is to determine the differences in bone health variables between NF1 patients and individuals without NF1, and between NF1 individuals without dystrophic scoliosis versus NF1 individuals who develop dystrophic scoliosis. Dual energy x-ray absorptiometry (DXA) and peripheral quantitative computerized tomography (pQCT) will be used to test the hypothesis that there-are subtle bone abnormalities of bone mineral density, bone area, bone mass, muscle-to-bone ratios, and cortical thickness in NF1. Urinary Dyridinium cross-links will be measured to detect differences in bone resorption. Spinal abnormalities in NF1 are not well understood, and dystrophic scoliosis is a highly morbid condition. This proposal will identify variables in patients with NF1 as prognostic factors for dystrophic scoliosis to improve clinical management.